Haptic Guidance for Robot Arm Teleoperation Using Ray-Based Holistic Collision Avoidance

We present a haptic guidance system for teleoperating a robot arm controlled by inverse kinematics. Unlike drones or vehicles, a robot arm occupies 3D space dynamically changing with its various configurations. With limited information on the remote environment and its current configuration, the remotely controlled robot arm has a higher chance of colliding with the surroundings. Consequently, users need to maintain a high level of attention, which results in fatigue during operation. In this paper, we propose a system of haptic force guidance that is robust to both the robot arm configuration and its surroundings. Our system first computes the guiding forces at multiple points in the robot arm using ray-based depth sampling. Then, haptic force feedback is generated by aggregating guiding forces using a motion-based approximate Jacobian. Our system requires minimal prior information about the environment and the robot arm. Moreover, the proposed ray-based depth sampling method is more efficient in computation time than the widely used potential field-based approach. User studies show that our system reduces the risk of collisions, as well as the mental workload during teleoperation in a virtual environment.